Session Overview
Session
GD7: Urban climatology studies II : tropical and arid climate cities
Time:
Thursday, 23/Jul/2015:
11:00am - 12:30pm

Session Chair: Lee Chapman, University of Birmingham
Location: St-Exupéry Amphitheater

Presentations

Investigating the Effect of Land Use/Land Cover on Urban Surface Temperature in Makurdi, Nigeria

Bernard Tarza Tyubee1, Raymond R. C. Anyadike2

1Benue State University, Nigeria; 2University of Nigeria, Nsukka, Nigeria

The study utilizes remote sensing, in conjunction with geographic information system (GIS), to explore the effect of LULC change on land surface temperature (ST) over 15-year period in Makurdi, North central Nigeria. A total of twelve (12) Landsat TM/ETM+ images are acquired for April, June and January of 1991, 1996, 2001 and 2006 for the study. The Landsat TM/ETM+ images are analyzed using Integrated Land and Water Information System (ILWIS) 3.3, ERDAS Imagine 8.6 and ArcGIS 9.3 software. The Normalized Difference Vegetation Index (NDVI), Normalized Difference Wetness Index (NDWI) and Normalized Difference Built-up Index (NDBI) are used to represent the dominant land use/land cover (LULC) types in the study area. In order to investigate changes in areas of ST, ST is grouped into seven (7) classes namely 27oC-29oC, 29oC-33oC, 33oC-37oC, 37oC-41oC, 41oC-45oC, 45oC-48oC and 48oC-51oC respectively. The effect of NDVI, NDWI and NDBI on ST is investigated using pixel-by-pixel correlation analysis. The results show that areas of water, forest, undergrowth/wetland and cultivated land have decreased by 4km2, 37km2, 119km2 and 19km2 from 1991-2006. Conversely the area of built-up land has increased by 179km2 during the same period. The areas of ST classes above 33oC (below 33oC) have increased (declined) by 249.1km2 (249.1km2) from 1991 to 2006. The ST is negatively correlated with NDVI and NDWI but positively correlated with NDBI for all the months/seasons and years. The results suggest that change in land use/land cover, driven by urbanization, is the primary driver of surface and atmospheric temperatures in Makurdi.

GD7-1-1501111_a.pdf

Spatial Distribution of Urban Heat Island and Intra-urban Air Temperature Variability in High-density Urban Areas in Hong Kong

Yingsheng ZHENG, Yuan SHI, Chao REN, Hung Lam, Steve YIM, Yuk Fo, Derrick LAI, Ka Lun,Kevin LAU

The Chinese University of Hong Kong, Hong Kong S.A.R. (China)

High level of urbanization in Hong Kong causes urban heat island effect (UHI), which is considered as one of the major urban climatic problems in recent years. Understanding the spatial distribution of UHI intensity as well as intra-urban air temperature differences is the basis of urban climatic planning for UHI mitigation. Many studies have investigated the impacts of urban surface properties and geographical characteristics on UHI intensity. However, given the unique urban morphology and complicated geographical environment in Hong Kong, the climatic conditions and UHI pattern at local scale exhibit greater diversity compared to other cities. Thus, there is a need to investigate the spatial pattern of UHI intensity, identify its relationships with urban morphological parameters, and propose planning strategies for UHI mitigation.

This paper firstly examines ten years of meteorological observations (2002- 2012) from Hong Kong Observatory (HKO) automatic weather stations (AWSs) to understand the diurnal urban heat island profile, and find out the linkages of urban heat island intensity with cloud amount and wind speed. Base on the data analysis, UHI intensities are found to be maximized during nighttime with calm wind and clear sky conditions. Secondly, this paper analyzes air temperature data captured by six mobile traverse measurements at street level (2.3m above ground) during calm and clear summer nights (8PM to 10PM) in 2014 to study the UHI intensity pattern over four high-density areas: Kowloon peninsula, north part of Hong Kong Island, Tuen Mun and Yuen Long. Thirdly, urban morphological parameters, such as building height, ground coverage ratio, sky view factor, aspect ratio, calculated using GIS are reported and compared with the traverse measurement data. The results indicate that, the compact high-rise commercial area Mong Kok (in Kowloon), Causeway Bay (in Hong Kong Island), Castle Peak Road (in Yuen Long) are the hotspot areas with higher UHI intensity of 3-5 °C. Intra-urban air temperature difference are identified to be 1.5-3.5 °C due to the variations in urban geometry and geographical conditions. Finally, the discussion part of this paper makes summaries about the spatial characteristics of nocturnal urban heat island effect as well as intra-urban air temperature differences, and also presents UHI mitigation strategies through optimization in land-use planning and building design.

GD7-2-1871100_a.pdf

STUDY OF HEAT ISLAND PHENOMENON IN ANDEAN COLOMBIAN TROPICAL CITY, CASE OF STUDY: MANIZALES, CALDAS -COLOMMBIA

Dalia Nuith Roncancio Rubio, Fernando Mejia

Universidad Nacional de Colombia sede Manizales, Colombia

The way the cities were built and their shapes, has effects on the incidence of solar radiation, remaining and heating the entire surface. Industrial activities also generate heat and emissions, which influence in the increased temperatures, changing the urban environment, favour the effect of a climatic phenomenon known as Urban Heat Island (UHI).

The Colombian cities, due to their topography, geographic ubication and anthropogenic activities can be affected for this phenomenon, making necessary and appropriate to develop studies of environmental urban temperature, to know the climatic dynamic and establish if exists or not the urban heat island in there.

In the case of Manizales ubicated in Caldas State, was developed UHI study, using the method of night thermal transects in 13 mobile stations, located at characteristic points of the city, used a pick-up truck in which, automatic mobile weather station Davis Instruments® was installed, belonging to the Environmental Studies Institute IDEA from National University of Colombia in Manizales headquarters.

The environmental temperature and wind speed data took at each point, making an altitude correction of sampled temperatures to develop thermal and wind speed maps. Also correlated these data with the temperature and wind speed of Environmental Studies Institute, IDEA network stations, although was calculated UHI intensity and temperature trend in the last eight years in the city.

The thermal maps, evidence that, environmental temperature of the city is concentrated in downtown of Manizales. The Pearson correlation between sampling temperature and wind speed data of network stations as well the UHI intensity and wind speed relation, was negative -0.119 and -0.114 respectively.

This indicates, the structural shape to the city of Manizales influence on the temperature rise in downtown, because there coincide with the commercial, the historical center, the political center (Caldas Office Governor and Mayor of Manizales), religious (Cathedral of Manizales) and Financial District, which influence in the high population density that goes around there daily and exerts a pressure on urban environmental factors, causing climatic environmental problems such as the heat island phenomenon.

The temperature trend in the last eight years has been the influence of the presence of El Niño phenomenon in 2010 and the La Niña in 2007, 2008 and 2009, years was affected by the phenomenon, produced invaluable economic damages in the Colombia Andean Region.

Key words: UHI, downtown, Temperature, wind speed.

GD7-3-1981188_a.pdf

Changing perspectives: Significance of long-term temperature observations in major cities

Alvin Varquez, Manabu Kanda

Tokyo Institute of Technology, Japan

Filtering out observation data from weather stations located in urban areas is a common procedure when estimating global surface temperature trends; urban stations are claimed to have little contribution to the estimated trends for large grids (>100-km grid spacing). Observations at urban stations can be likened to noise in grid-estimated trends due to the rapid land surface changes in urban areas. The disregard of surface temperature trends in urban areas can cause society to underestimate the significance of their own local climate. This study aims to investigate the trend differences of surface temperature of weather stations and the larger grid to which it belongs. The Berkeley Earth Surface Temperature (BEST) gridded data, along with the quality controlled urban weather stations, were utilized. Stations within cities in Southeast Asia, East Asia, and the Middle East with available long-term observation records were selected. Trends of monthly values of the minimum temperature, maximum temperature and the average temperature were acquired from linear regression using least squares fitting. The period covered was from the 1960s to the present. Subtracting the trend of an urban weather station and its encompassing grid allows the estimation of the trend of urban heat island (UHI). Results show that surface temperature has been rising faster in urban areas than the grid-estimated surface temperature especially in Southeast and East Asian cities. Similar to global temperature trends, UHI trend estimated from minimum monthly temperature (minimum temperature twice than that of the grid-estimated values) was found to be larger than the UHI trend estimated from average and maximum monthly temperature. UHI trend estimated from maximum temperature was also found to be positive but much less than UHI trend estimated from minimum temperature. The differences in the trend of UHI per continent suggest that UHI depends not only on the rate and level of urbanization but also to the climate zone to which it belongs. Ranking the UHI trend estimates, East Asian cities such as in Japan and South Korea have slightly more rapid increase in UHI (>0.2 to 0.6 K/100 years) than Southeast Asian cities. Interestingly, observed long-term temperatures in Middle Eastern cities, which are mostly arid, revealed lesser or insignificant UHI trend. With the increase in urban-rural population ratio and GDP in developing countries of Southeast Asia, the trend in UHI can be a critical parameter aside from the trend of global surface temperature.

GD7-4-4451149_a.pdf

Exploring the Spatial and Temporal Variation of Air Temperature in the Extreme Desert Climate of Doha, Qatar

M Salim Ferwati1, Yasuyo Makido2, Vivek Shandas2, David Sailor2, Dylan BothamMyint2

1Qatar University, Qatar; 2Portland State University, USA

The question of mitigating extreme urban heat takes on a new dimension and challenge in a desert climate such as Doha, Qatar. Throughout much of the year the daytime air temperatures remain unbearable, limiting outdoor activities to late evening strolls along the Corniche waterfront promenade. Urban heat mitigation in such a climate will have little effect on the tolerability of daytime extreme temperatures, but may have the potential to increase accessibility to moderate outdoor conditions in the early morning and late evening hours. A first step in estimating the potential for mitigating heat in such an extreme desert climate is to understand the spatial and temporal variation of air temperature throughout the city. Once this understanding is in place, this knowledge can be combined with information on spatial variations in surface characteristics and activity patterns in an attempt to develop a predictive model that relates these physical characteristics to the corresponding variations in near-surface air temperatures.

To address this challenge we have established a network of rooftop weather stations distributed across the city. These stations have recorded conditions at 15 minute intervals since September 2014. These data have been combined with a spatially denser array of vehicle-based traverse measurements and available parcel-level land cover data to develop an initial assessment of the spatial and temporal variability of air temperature in this desert city and prospects for mitigation. The resulting analysis will provide insight into the role of local land cover in temperature variations across the city, leading to the development of specific recommendations for future development in the region.

GD7-5-5811483_a.pdf

Analysis of observed temperature trends over urban, town and rural areas of Pakistan

Sajjad Hussain SAJJAD1,2, Nadège BLOND1,2, Manon KOHLER2, Alain CLAPPIER1,2

1University of Sargodha, Sargodha - PAKISTAN, Pakistan; 2Laboratoire Image Ville et Environnement, Université de Strasbourg, France

The purpose of this work is to study the evolution of temperatures at several locations in Pakistan. A comparative quantification of the changes in temperatures on urban, town and rural areas of Pakistan has been done. For this purpose, daily mean (Tm) minimum (Tn) and maximum (Tx) temperatures data averaged on an annual, monthly and seasonal basis from 1950 to 2004 of 42 stations were obtained from Pakistan Meteorological Department (PMD). The data was homogenized by using Standard Normal Homogeneity Test (SNHT). The resulting homogenized data was analyzed by using least square linear regression for two different periods: 1950–1979 and 1980–2004. The analysis shows that the annual mean, minimum and maximum temperatures over Pakistan are increasing. The trends of annual mean, minimum and maximum temperature observed over urban, town and rural stations during 1980–2004 are significantly higher than the trends observed during 1950–1979. A higher growth in minimum and maximum temperature over urban and town areas is observed. The increase in minimum temperatures is more important on urban areas than on town areas, while the maximum temperatures increase more on town areas than on urban areas. The tendencies are less important in summer than other seasons of the year.